Olefinic hydrocarbons, such as ethylene, propene, butene, and isobutene, are critical intermediates in the petrochemical industry. In order to satisfy market demand, substantial efforts have been invested in the production of such compounds by conventional thermal steam cracking of alkanes and naphtha and by catalytic dehydrogenation methods. However, conventional steam cracking is equilibrium limited and requires very high temperatures (over 700° C.) to achieve a high enough conversion of ethane to be economically viable. Even so, temperatures are limited by available alloys to temperatures at which single pass yields are still relatively low. Steam cracking also requires the input of large amounts of heat to drive the endothermic gas phase cracking reaction. Because of the equilibrium limitations, steam cracking must be carried out at low pressures typically 1 atmosphere or less and requires cooling and compression of the product stream to effect separation and recovery of the olefins produced.
Conventional catalytic dehydrogenation has similar disadvantages, including the need for high reaction temperatures (e.g., 550 to over 700° C. depending on the feedstock), the deactivation of the catalyst by coke formation, and the consequent need for continuous or periodic catalyst regeneration at frequent intervals throughout the process. In addition, there are thermodynamic limitations in conventional dehydrogenation. These thermodynamic limitations are due to the fact that conversion in conventional dehydrogenation processes are equilibrium limited, and require high temperature and low pressure to achieve high single pass yields. As a result of these substantial drawbacks, the petroleum industry has sought a solution to the demand for olefinic hydrocarbons in the use of autothermal cracking and oxidative dehydrogenation methods.
In autothermal cracking, oxygen or air is added to the feed and partially combusts part of the feed in situ generating the high temperatures required to thermally crack the remaining feedstock. In some variants a catalyst is used to support combustion with the catalyst being in the form of a fixed bed or a fluidized or spouted bed. Fixed beds are preferred to reduce catalyst attrition. In some cases hydrogen is co-fed with the feedstock and is found to increase olefin yields. Autothermal cracking usually takes place at high temperatures (550-1200° C.) and requires very short reaction times and rapid quenching of the products to preserve the olefinic products and prevent further undesirable reactions. Even so, by products are formed including carbon oxides. At higher pressure, yields of undesirable by-products increase. At very high temperatures as encountered in some autothermal processes, hydrocarbon cracking to methane also reduces selectivity to useful olefinic products.
Catalytic oxidative dehydrogenation is, in principle, not subject to many of the problems associated with conventional steam cracking or catalytic dehydrogenation because of the presence of oxygen in the reaction mixture. Oxidative dehydrogenation (ODH) uses oxygen to react with the hydrogen released from the hydrocarbon, in situ, so that the aforementioned equilibrium limitation is removed, and high single pass yields can be achieved. The reaction is exothermic overall and does not require a supply of heat as in endothermic dehydrogenation reactions. Generally, in a catalytic oxidative dehydrogenation process, the reactants (hydrocarbon and an oxygen-containing gas) are passed over the fixed bed catalyst directly to produce olefin product. Typically, the hydrocarbon is a saturated hydrocarbon such as ethane or a mixture of saturated hydrocarbons. The hydrocarbon may be gaseous or liquid at ambient temperature and pressure but is typically gaseous.
An example of an alkene which can be formed via an oxidative dehydrogenation process, is ethylene. The latter process is attractive for many reasons. For example, compared to thermal cracking, high ethane conversion can be achieved at moderate temperatures (300-1000° C.) by catalytic oxidative dehydogenation. Unlike thermal cracking and catalytic dehydrogenation, catalytic ODH is exothermic, requiring no additional heat, beyond feed pre-heat, to sustain reaction. Furthermore, in contrast to catalytic dehydrogenation, catalyst deactivation by coke formation should be minimal in ODH because of the presence of oxygen in the reactor feed. Other alkanes can similarly be oxidatively dehydrogenated.
Although there are no reported commercial ODH processes operating at the present time, there is a high level of commercial interest. Activity has focused on ethane, propane and isobutane ODH, and patents to same have issued. Representative of these patents are the following US patents, all of which are herein incorporated by reference: U.S. Pat. No. 4,524,236; U.S. Pat. No. 5,162,578; U.S. Pat. No. 5,593,935; U.S. Pat. No. 5,997,826; U.S. Pat. No. 6,313,063; U.S. Pat. No. 6,281,378; U.S. Pat. No. 6,239,325; U.S. Pat. No. 6,235,678; U.S. Pat. No. 6,130,183; U.S. Pat. No. 6,355,854 and U.S. Pat. No. 6,310,241.
Industrial interest has stimulated investigations into new catalysts and methods for improved performance (e.g., conversion and selectivity) for the oxidative dehydrogenation of alkanes. U.S. Pat. No. 4,524,236 reports high conversion (73%) and high selectivity (71%) for ethane ODH but these results were obtained only by diluting the ethane/oxygen feed with helium as 85.5% of the feedstock. Others have achieved high yields by co-feeding hydrogen with the hydrocarbon feedstock and oxygen (see U.S. Pat. No. 5,997,826).
In U.S. Pat. No. 4,524,236 McCain describes a process for the low temperature catalytic oxydehydrogenation of ethane to ethylene in a gas phase and featuring the use of a catalyst containing Mo/V/Nb/Sb and an additional element.
There have been different approaches to adding oxygen to the ODH reaction. Lodeng et al. in U.S. Pat. No. 5,997,826 describes a process for converting C3 and C4 paraffins to olefins by a sequential reactor that contains at least three zones, a catalytic dehydrogenation process zone, an oxygen admixing zone,and a catalytic oxidation zone, wherein the flow velocity in the admixing zone is higher than in the catalyst zones. Ward in U.S. Pat. No. 4,739,124 discloses mixing oxygen between stages.
In a process for catalytic selective oxidation of a hydrocarbon, Perregaard et al. in U.S. Pat. No. 6,515,146 discloses a reactor in which oxygen flows into a 7 mm inner diameter tube through the porous alumina tube walls and into the catalyst bed held within the tubes. No mention is made of the useful of this approach in ODH.
Beretta et al. in “Production of olefins via oxidative dehydrogenation of light paraffins at short contact times,” Catalysis Today, 64 pp103-111 (2001) reported testing of a Pt/Al2O3/Fe—Cr catalyst in an annular reactor. Comparative tests without catalyst showed no proof that the Pt catalyst contributed to the selective oxidation of ethane to ethene; however, there was strong proof “that the catalyst was active in non-selective oxidation reactions, and that gas-phase oxidative pyrolysis was a fast process with very high ethene selectivities.” The authors concluded that the Pt-containing catalyst seemed to be mainly active in the total oxidation of ethane to COx.
Several workers have described oxidative dehydrogenation in catalyst monoliths positioned in conventional reactors. See, U.S. Pat. Nos. 4,940,826, 6,166,283, and 6,365,543. They do not suggest the use of monoliths in microchannel reactors or any microchannel advantages.
As compared to conventional, fixed bed reactors, microchannel reactors have been found to suppress thermal gradients; however, at comparable catalyst bed temperatures, the microchannel reactor did not improve performance. Steinfeldt et al. in “Comparative Studies of the Oxidative Dehydrogenation of Propane in Micro-Channels Reactor Module and Fixed-Bed Reactor,” Studies in Surface Science and Catalysis, pp185-190 (2001) conducted testing of ODH in a microchannel reactor over a VOx/Al2O3 catalyst. To minimize temperature gradients, the catalyst was diluted with quartz in a ratio of 1:9. The authors reported that “the use of microchannels reactor module allowed isothermal operation at all reaction conditions.” The authors concluded that the “micro-channel reactor module and fixed bed reactor show approximately the same catalytic results under isothermal conditions.”
Despite extensive research, there remains a need for new oxidative dehydrogenation catalysts, catalytic systems, and methods that achieve high conversion at high selectivity, such that the yield of the desired olefin is maximized, and extraneous oxidative side reactions are minimized. Such extraneous oxidative side reactions may include the conversion of starting hydrocarbon, e.g., alkane, into carbon oxides (CO and/or CO2), and/or conversion of desired product alkene into carbon oxides.